1922] 



PAYSON — STUDY OF THELYPODIUM AND ITS IMMEDIATE ALLIES 235 



of these remarks to demonstrate the general characteristics of 

 the common ancestor, the phylogenetic sequence of characters, 

 and to correlate these findings with the geographical distri- 

 bution of the various units. 



The gynophore. — Thelypodium received its name because of 

 a gynophore or stipe that in many species raises the ovary and 

 fruit above the torus. Of the fourteen species in the genus, seven 

 are found to possess a distinct stipe and in the others the ovary 

 is not quite sessile. Five species of the seven show a stipe that 

 at least occasionally reaches a length of two millimeters. It is 

 most highly developed in T. laciniatum and T. eucosum, being 

 usually more than two millimeters long. What is the significance 

 of this stipe? 



Recent students of the Cruciferae are nearly unanimous in the 

 belief that this family has been derived from the Capparidaceae 

 or from capparidaceous-like ancestors. In this connection it is 

 sufficient to refer to the writings of Hallier, 1 Lotsy, 2 von Hayek, 8 

 and Engler and Gilg. 4 In the Capparidaceae a stipe is almost 

 universally present and is often very long. Granting this rela- 

 tionship between the two families, the presence of a stipe in the 

 Cruciferae, in which it is not of common occurrence, must be 

 considered either a primitive character or an atavistic variation. 

 For the sake of argument and as a point of attack the stipe in 

 this group will be assumed to be a primitive characteristic, 

 question now arises : Are the two species that possess the loi 

 stipe primitive in other respects? 



The pods. — In many genera of the Cruciferae the capsul 

 pods are known to display characters of the utmost imporl 



The 



taxonomically. This is not true of Thelypodium to such an ex- 

 treme degree, except as regards the cellular structure of the sep- 

 tum. The capsules are often particularly variable in length 

 They are usually terete, glabrous, and many times longer thai 



wide. Frequently the valves 

 the seeds and the pod becomes 



re somewhat constricted between 



torulose in consequence. In two 



,ffine, the valves are slightly com- 



pressed parallel to the septum. It has been found impossible 



much 



this character within 



UTallier, H. Provisional scheme of the natural (phylogenetic) system of 



flowering plants. New Phytol. 4:157. 1905. 



Mjotsy, J. P. Vortrage iiber botanische Stammcsgoschichto 3:915-916. 1911. 



"von Hayek. Entwurf eines Crueiferen-Systems auf phylogenctiseher Grund- 



lage. Beih. Bot. Centralbl. 27 1 : 176-178. 1911. 



'Engler, A. und Gilg, E. Syllabus der Pflanzenfamilien, 8th ed., 201. 1919. 



